During normal development, a switch from the production of fetal hemoglobin (containing gamma chains) to the production of adult hemoglobin (containing beta chains) occurs. The gamma to beta switch in the synthesis of hemoglobins is not well understood. Hydroxyurea is used in the treatment of sickle cell anemia to increase the relative amount of fetal to sickle hemoglobin. Studies of hemoglobin switching and of drugs which affect it, such as hydroxyurea, have been difficult due to the absence of a culture system in which erythroid progenitors could be isolated in sufficient numbers to perform molecular studies. We have used a recently developed erythroid culture system (Fibach, 1989) to study the effect of hydroxyurea in vitro. Using HPLC, we have been able to show an apparent increase in fetal hemoglobin per cell as well as in per cent fetal hemoglobin. Total hemoglobin per cell also appears to increase in thalassemic cultured cells. These results are similar to those found in vivo where treatment of sickle cell patients with hydroxyurea results in increases in percent fetal hemoglobin, fetal hemoglobin per cell and total hemoglobin per cell. We have been able to demonstrate that the effect of hydroxyurea on erythroid cells is occurring at a relatively late stage in erythroid maturation while the cytotoxicity of hydroxyurea toxicity is greatest at the earlier maturational stages, prior to the onset of obvious hemoglobin formation. Also we have shown that hydroxyurea seems to increase the absolute amount of a modified hemoglobin. The significance of this finding is uncertain at this time. Increasing erythropoietin did not result in increased fetal hemoglobin. We believe that this non-neoplastic primary erythroid culture system will be a useful experimental system in which to examine in detail the molecular and cellular mechanism(s) of pharmacological induction of Hb F.